Optimizing the Hachinski ischemic scale

Optimizing the Hachinski ischemic scale

S350 P2-124 Poster Presentations P2 CLINICAL ASSESSMENT OF ASIANS WITH SUBJECTIVE MEMORY COMPLAINTS: THE CHINESE MINI-MENTAL STATE EXAMINATION LACKS ...

58KB Sizes 37 Downloads 116 Views

S350 P2-124

Poster Presentations P2 CLINICAL ASSESSMENT OF ASIANS WITH SUBJECTIVE MEMORY COMPLAINTS: THE CHINESE MINI-MENTAL STATE EXAMINATION LACKS SENSITIVITY IN THE MILD END OF THE DISEASE SPECTRUM

Mark Chan, Ying Ying Yeo, Mei Sian Chong, Tan Tock Seng Hospital Singapore, Singapore, Singapore. Contact e-mail: peng_chew_chan@ttsh. com.sg Background: With increasing emphasis on the early identification of cognitive impairment, it is imperative that screening tools are evaluated for their discriminating properties in the mild end of the disease spectrum. We aimed to assess the validity and diagnostic utility of the Chinese Mini-Mental State Examination (CMMSE) when applied to subjects with early cognitive complaints in an Asian memory clinic setting. Methods: 206 subjects with no dementia (ND, n ¼ 44), mild cognitive impairment (MCI, n ¼ 50), and mild dementia (n ¼ 112) presenting over a 2.5year period were reviewed. MCI subjects were operationally defined as global Clinical Dementia Rating (CDR) score of 0.5, at least one domain >1SD impairment upon neuropsychological testing and functional abilities preserved while mild dementia subjects had a global CDR of 0.5 or 1.0 and fulfilled standardized criteria for dementia diagnosis. Correlations of CMMSE with global {CDR-sum of boxes score (CDR-SB)}, functional {Lawton’s scale for instrumental activities of daily living (iADL)} and neuropsychological (neuropsychological test battery composite score) measures were examined, while receiver operating characteristic (ROC) curves were used to determine optimal cut-offs and corresponding sensitivity (Sn), specificity (Sp) and area-under-curve (AUC). Results: Subjects were predominantly elderly (mean age ¼ 72.5 6 8.6 years), female (55.3%) and Chinese (88.8%). CMMSE had significant but modest correlations with CDR-SB, iADL and neuropsychological test performance (Pearson’s R ¼ -0.61, 0.47 and 0.55 respectively, all p < 0.001). The AUC, Sn and Sp for the entire cohort were 0.85, 75.0% and 80.9% respectively, with an optimum cut-off score to exclude dementia of 25. The best test performance were obtained for subjects with higher educational attainment (AUC, Sn and Sp ¼ 0.87, 83.3% and 80.3% respectively), and worst when only ND and MCI subjects with less than 6 years of education were considered (0.67, 64.3% and 73.3% respectively). Conclusions: Despite higher cut-offs, the sensitivity and specificity of CMMSE for the detection of MCI and mild dementia were substantially lower than previously reported values based on the evaluation of mild dementia alone. Screening of subjects with early cognitive complaints using the CMMSE must be supplemented with neuropsychological testing and other biomarkers to augment its diagnostic yield. P2-125

PRESENCE OF EXTRAPYRAMIDAL MOTOR SIGNS PREDICT WORSE COGNITIVE AND FUNCTIONAL TRAJECTORY OF DECLINE IN MILD-TOMODERATE ALZHEIMER’S

Alireza Atri1,2, Jessica S. Dodd1,2, Frances M. Yang3,4, Joseph J. Locascio1, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA; 2GRECC, VA Medical Center, Bedford, MA, USA; 3Beth Israel Hospital, Boston, MA, USA; 4Institute for Aging Research, Hebrew Senior Life, Boston, MA, USA. Contact e-mail: [email protected]

assessed using mixed effects models covarying for baseline scores on the 30-point MGH-EPS severity scale and factor scores. Results: Clinical trajectory significantly differed between patients with and without EPS. After adjusting for baseline scores, age, education, and duration of illness, the EPS group had a significantly faster cognitive decline (p ¼ .025). ADL decline was worse for the EPS-at-baseline group, then for the EPS-not-baseline, and least for the NO-EPS group (p ¼ .012). Using EPS severity as a time varying visit-level predictor, those with worse scores on the MGH-EPS scale (Cronbach a ¼ 0.77) had worse BDS and ADL scores (p < .0001). Factor scores (level of EPS severity) at baseline predicted future ADL, but not BDS scores (p ¼ 0.01). Conclusions: EPS is associated with significantly worse cognitive and functional decline in AD. Study strengths include a large, well-defined, prospectively collected, single-site cohort study utilizing validated measures of cognition and function, and mixed effects models. The results suggest that presence or development of EPS in the course of mild or moderate AD, likely due to more widespread AD neuropathology, may assist clinicians in prognostication and in estimating progression of clinical course. They also lend support to the premise that EPS severity should be assessed in AD clinical trials, and should be accounted for in analyses as a potential covariate that may affect risk stratification and rate of clinical progression. P2-126

RELATIVES OR PATIENTS: WHO BETTER DETECT THE MEMORY IMPAIRMENT?

Edla Da Silva, Lı´via Gonc¸alves Rodrigues, Renata Kochhann, Cla´udia Godinho, Diego Onyszko, Andre´a Heisler, Maria Otı´lia Cerveira, Ana Luiza Camozzato, Ma´rcia Lorena Chaves, Hospital de Clı´nicas de Porto Alegre, Porto Alegre, Brazil. Contact e-mail: [email protected] Background: Besides age, other factors have been studied as candidates for predictors of cognitive impairment. The aim of this study was to evaluate the association of subjective memory complaints with positive screening on the Mini Mental State Examination (MMSE) in outpatient elders. Methods: The Mini Mental State Examination was applied to 618 outpatient elders awaiting non-memory related consultations, in three medical specialties (Internal Medicine, Cardiology and Rheumatology) in a University hospital of a major southern city of Brazil. Cognitive dysfunction was defined as MMSE 24 and education  5 years or MMSE 17 and education  4 years. Subjective memory complaints reported by patients as well as by their relatives were also recorded. We also analyzed the following characteristics: sex, age, education, medical specialty, and whether the patient was escorted by relatives at the consultation. Results: Age, subjective memory complaints reported by patients and complaints reported by relatives, and escorted patients were associated with cognitive impairment (p ¼ 0.007, p < 001, p ¼ 0.001, p ¼ 0.001, respectively) in the univariate analysis. In the logistic regression model, only subjective memory complaint reported by relatives was associated with cognitive impairment (OR ¼ 2.7; p ¼ 0.007). Conclusions: The memory impairment complaint reported by relatives was strongly associated with cognitive impairment in elderly patients.

1

Background: Previous research has suggested that AD patients with extrapyramidal motor signs (EPS) may decline more rapidly. However, discrepancies remain as to whether and how much EPS affects function, cognition, or both. Our objective was to assess, in a prospective AD clinical cohort, the impact of EPS on long-term course of cognition and function. Methods: 1,883 patients with PrAD (mean age ¼ 74.7, education ¼ 13.6 years) underwent several evaluations at the MGH Memory Disorders Unit. Patients were divided into 3 groups based on the presence of motor impairments (EPS-at-baseline, EPS-not-baseline, and NO-EPS). EPS was present if dysarthria, Myerson’s sign, akinesia, abnormal tone, action tremor, dystonia, abnormal arm swing, and/or a parkinsonian gait was recorded. Cognition and function were measured by the Blessed Dementia (BDS) and Weintraub ADL scales, respectively. Impact of EPS severity on clinical course was

P2-127

OPTIMIZING THE HACHINSKI ISCHEMIC SCALE

William R. Shankle1,2, Shahram Oveisgharan3, Vladimir Hachinski3, 1 Medical Care Corporation, Irvine, CA, USA; 2Department of Cognitive Science, University of California at Irvine, Irvine, CA, USA; 3Department of Clinical Neurological Sciences, University Hospital, University of Western Ontario, London, ON, Canada. Contact e-mail: [email protected] Background: The Hachinski Ischemic Scale (HIS) is a valid instrument that differentiates cerebrovascular (VD) vs. non-cerebrovascular (non-VD) dementing disorders. Proper scaling theory has not been applied to the HIS to determine the correct scaling of its item responses and the number of independent dimensions it captures. Addressing these issues could improve accuracy and shorten the instrument. Methods: 2,968 subjects in the Canadian Study for Health and Aging had complete HIS responses. 702 subjects had either VD (N ¼ 192) or non-VD (N ¼ 512) diagnoses. HIS items with low frequencies were merged with their most similar HIS item to create composite HIS items to avoid over-fitting the data. Correspondence analysis

Poster Presentations P2 (CA) was applied to the HIS data on all 2,968 subjects to generate optimally scaled item responses and patient scores. CA was performed separately on different configurations of the HIS_1) the original 13 items, subsets of items, and subsets using composite items. We used the optimal or raw HIS item scores plus covariates of age, sex, education, history of hypertension and diabetes to predict VD vs. non-VD diagnosis. Using bootstrap sampling, we generated 100 data sets of an equal number of randomly sampled VD and non-VD subjects, and then performed logistic regression on each data set to obtain stable parameter estimates. Performance statistics were determined using non-parametric receiver operating characteristic curve methods. Results: Two dimensions of the optimally scaled HIS item responses, and none of the covariates, significantly contributed to VD/non-VD prediction. Compared to the original, 13-item HIS instrument scoring, an optimally scored, composite item model consisting of 5 questions was more accurate (96.5 vs. 97.3%), more sensitive (91.7% vs. 92.2%) and more specific (89.6% vs. 92.1%). The best, optimally scored subset of 6 HIS items performed comparable to the original HIS instrument. Conclusions: 1) The HIS instrument can be substantially shortened and improved by optimal scaling and by creating composite items. 2) The HIS instrument measures two dimensions of information. 3) The composite HIS item model does not require a neurological exam to accurately classify VD and non-VD. P2-128

CARDIOVASCULAR FACTORS AND WHITE MATTER LESIONS AFFECT ON EXTRAPYRAMIDAL SIGNS IN ALZHEIMER’S DISEASE

Moon Ho Park1, Duk L. Na2, Hae Ri Na3 CRCD Study Group1Korea University Medical College, Ansan Hospital, Ansan-si, Republic of Korea; 2 Sungkyunkwan University, Seoul, Republic of Korea; 3Bobath Memorial Hospital, Seongnam-city, Republic of Korea. Contact e-mail: parkmuno@ yahoo.co.kr Background: Extrapyramidal signs (EPSs), which are important characteristics of Parkinson’s disease (PD), occur frequently in Alzheimer’s disease (AD). Although AD as well as PD share common clinical features such as EPSs, these diseases vary with respect to cardiovascular factors. The presence of cardiovascular factors increases the risk of AD is; however, these factors have been known to be inversely associated with PD. The objective of this study was to assess the effect of cardiovascular factors on EPSs in AD and their relation to the overall AD risk. Methods: We recruited 1,187 consecutive subjects with clinically probable AD and 333 subjects with no cognitive impairment and without EPSs (control) from the Clinical Research Center for Dementia (CRCD) using a registration system under the name of Clinical Research for Dementia of South Korea (CREDOS). All participants underwent detailed clinical evaluations which included assessments of cardiovascular factors, cognitive function, and EPSs, as well as white matter lesions (WMLs) on brain MRIs. Extrapyramidal sign subtypes were classified into tremor-dominant (TD), postural instability gait difficulty (PIGD), or indeterminate, and WML subtypes were classified into periventricular WMLs (PWM) or deep WMLs (DML). Results: EPSs were present in 17.9% of subjects with AD and were associated with classical cardiovascular factors such as age, male gender, diabetes mellitus, and WMLs. Additionally, a multivariate logistic regression analysis determined that EPSs in AD were associated with PWM (odd ratio [OR], 1.61-2.52). With respect to EPS subtype, the majority (78.4%) of EPSs in AD were PIGD, which was also associated with WMLs (OR 1.84-2.41), PWM (OR 2.09-3.14), and DWM (OR 1.83-3.42). Conclusions: EPSs in AD are associated with selected cardiovascular factors as well as WMLs. P2-129

PREDICTING MILD COGNITIVE IMPAIRMENT STATUS WITH MARKERS OF COGNITIVE AND PHYSIOLOGICAL CHANGE

Anna Braslavsky1, Roger A. Dixon2, Holly Tuokko1, Stuart W. S. MacDonald1, 1University of Victoria, Victoria, BC, Canada; 2 University of Alberta, Edmonton, AB, Canada. Contact e-mail: abraslav@ uvic.ca

S351

Background: Mild cognitive impairment (MCI), postulated as a prodromal stage of dementia, may be predicted by previous changes in cognitive and biological health status. This study explores the use of markers of cognitive and physiological change in predicting MCI classification. Methods: Participants were a sample of community-dwelling older adults from the Victoria Longitudinal Study (VLS), a longitudinal sequential study of aging. Participants were from Waves 5, 6 and 7 of Sample 1 (n ¼ 129, mean age at last testing ¼ 85.38), Waves 3, 4 and 5 of Sample 2 (n ¼ 309, mean age ¼ 83.27) and Waves 1 and 2 of Sample 3 (n ¼ 577, mean age ¼ 72.07). At final testing, participants were stratified into 4 cognitive status groups based on age and education: Young-old (YO; low education) n ¼ 79, YO (high education) n ¼ 382, Old-old (OO; low education) n ¼ 30 and OO (high education) n ¼ 81. Group means were calculated for 5 cognitive reference domains (perceptual speed, inductive reasoning, episodic memory, verbal fluency, and vocabulary). Participants scoring 1 or more standard deviations (SD) below the group mean on episodic memory were classified as single-domain MCI. Participants 1 or more SD below the mean on episodic memory and at least one other measure were classified as multiple-domain MCI. Remaining participants were classified as non-impaired. Results: Linear mixed models were used to identify individual rates of change, spanning up to a 6-year interval, for measures of neurocognitive speed, semantic memory, peak expiratory flow and blood pressure. Within-person slopes were derived for each measure and used to predict MCI classification. Results indicate an increased risk of single- and multiple-domain MCI status as a function of decline in fact recall (MCI-S: p ¼ 0.005, OR ¼ 1.641; MCI-M: p ¼ 0.000, OR ¼ 1.821) and reading span (MCI-S: p ¼ 0.001, OR ¼ 1.999; MCI-M: p ¼ 0.000, OR ¼ 1.999). Changes in computation span were associated with increased risk of multiple-domain MCI (p ¼ 0.011, OR ¼ 1.988). Contrary to expectations, changes on physiological measures were not predictive of cognitive status over and above measures of cognitive change. Conclusions: Cognitive change in fact recall, reading span and computation span has potential clinical applicability for screening of cognitive status. Future research should continue to assess the utility of cognitive change and non-invasive biomarkers for predicting stability of MCI classification over time. P2-130

ANXIETY IS ASSOCIATED WITH MILD COGNITIVE IMPAIRMENT: THE MAYO CLINIC STUDY OF AGING

Andrea A. Obermeier, Yonas E. Geda, Rosebud O. Roberts, David S. Knopman, Teresa J. H. Christianson, V. Shane Pankratz, Bradley F. Boeve, Walter A. Rocca, Ronald C. Petersen, Mayo Clinic, Rochester, MN, USA. Contact e-mail: [email protected] Background: Little is known about the prevalence of anxiety in mild cognitive impairment (MCI). Therefore, we investigated the association of anxiety with MCI in a population-based setting. Methods: We conducted a casecontrol study derived from the population-based Mayo Clinic Study of Aging in Olmsted County, Minnesota. A random sample of 1,962 elderly participants aged 70-90 years (n ¼ 1,635 cognitively normal persons and n ¼ 327 subjects with MCI) constituted the sample for this study. We measured the prevalence of anxiety using the Beck Anxiety Inventory (BAI) and used a cut-off score of >9. A score of 0-21 indicates low anxiety; a score of 22-35 indicates moderate anxiety; a score of 36-63 indicates a high level of anxiety. We then compared the prevalence of anxiety in subjects with MCI to the prevalence in cognitively normal persons. Multivariable logistic regression analyses were conducted to compute odds ratios (OR) and 95% confidence intervals (CI). Results: Among 1,635 cognitively normal individuals, the median age was 79.6 years (range, 70.5. - 91.1 years); 827 (50.6%) of them were women. The BAI score was >9 in 115 individuals (7.0%). Among the 327 subjects with MCI, the median age was 82.7 (range, 70.9 - 91.8 years); 135 (41.3%) of them were women. The BAI score was >9 in 41 individuals (12.5%). After adjusting for age (continuous), sex, education (continuous in years), and Charlson Index (continuous), anxiety was significantly more prevalent in MCI than among cognitively normal persons (OR [95% CI] ¼ 1.69 [1.14. 2.51]; p ¼ 0.008). Conclusions: In a population-based setting, the prevalence of anxiety is significantly higher in MCI than in